SupplementaryInformation Figure S1 Capacity for liver regeneration declines with age. 3-month-old (red), 10–12-month-old (blue) and >18-month-old (green) nonpregnant female mice were subjected to 2/3 partial hepatectomy. In each mouse, liver volume was determined by MRI on the indicated days and recorded as a percentage of the liver volume before partial hepatectomy (mean ± s.e.m.). Note that while the age effect may seem to be transient, it results in considerable mortality.

Supplementary Information Figure S2 Improved recovery of liver function in pregnant mice after partial hepatectomy.10-12 month old pregnant and nonpregnant mice were subjected to partial hepatectomy. a. Blood was collected 24 h after partial hepatectomy and prothrombin time was measured (mean ± s.e.m., n = 3, P < 0.05, Kruskal-Wallis test). Prothrombin time values in nonhepatectomized controls ranged between 11.0 and 12.3 seconds (dashed line). b. Locomotor activity of mice in the indicated groups, 1 day after partial hepatectomy, was monitored by an open-field recorder as detailed in the materials and methods section (n ≥ 4; P < 0.0001, Mann-Whitney test).

SupplementaryInformation Figure S3 BrdU incorporation in the bowel is not affected by pregnancy. Representative immunohystochemical staining for BrdU of small bowels from the indicated groups. BrdU was administered continuously in the drinking water from the time of partial hepatectomy until sacrifice four days later.

SupplementaryInformation Figure S4 Hypertrophy of hepatocytes in hepatectomized pregnant mice. a. Aged pregnant and nonpregnant mice were subjected to partial hepatectomy. Four days after surgery mice were re-anesthetized and single-cell suspensions of isolated hepatocytes were prepared. Forward scatter values were determined for each preparation. b. Hepatocytes (4 × 105) from pregnant and nonpregnant mice were resuspended in 50 l of PBS, loaded onto a hematocrit capillary, and centrifuged at 3000g. c. Mean hepatocyte volume was calculated for each mouse by measuring the total volume of cells and dividing by cell number.

SupplementaryInformation Figure S5 Hepatocyte proliferation after delivery. Aged pregnant mice were hepatectomized at near term (n = 5) or left untreated (n = 4). After delivery, BrdU was administered through the drinking water for 6 weeks. The mice were then sacrificed and their livers were examined histologically. Hepatocyte proliferation, indicated by BrdU incorporation, was determined by immunohistochemistry. Shown are mean proliferation values ± s.e.m. and representative photomicrographs of BrdU-stained liver sections.

SupplementaryInformation Figure S6 Following hepatectomy, p53 and its target p21 are upregulated in nonpregnant mice. Liver sections from mice at the indicated days after hepatectomy were immunostained for p53 or p21. The extent of positive nuclei was assessed by two observers that were blinded to the treatment group. Values are mean ± s.e.m.

SupplementaryInformation Figure S7 Activation of the Akt/mTORC signaling pathway. Western blot analyses of liver extracts from aged nonpregnant and pregnant mice,2 days after 2/3 partial hepatectomy.

SupplementaryInformation Figure S8 Cell-size distribution, 4 days after surgery, in hepatectomized livers of young untreated (blue) and bpV(phen)-treated (orange) nonpregnant mice. Each data point is representative of at least three mice.

Supplementary Materials and Methods

Animal studies and tissue preparation

Mice aged 18 months or older ('old' mice) were either purchased from the National Institute of Aging and from Charles River Laboratories or maintained up to the required agein the Specific Pathogen Free (SPF) animal facility at our institution. Mice aged 10–12 months ('aged' mice) were purchased from Harlan Laboratories. Progesterone levels were measured to verify pseudopregnancy induction(Shiotani et al. 1993). BrdU (100l/10g body weight; Amersham) was injected i.p. at the indicated times before the mice were killed. When indicated, BrdU (#B5002; Sigma) was added to the drinking water (0.8 mg/ml). The mice were allowed to drink ad libitum. For all experiments mice were killed by cervical dislocation. In some cases, a liver sample was removed and 'snap-frozen' for protein and RNA analyses. Livers were removed, weighed, photographed, and fixed in formalin overnight, and the next day the entire liver was embedded in paraffin. For measurement of serum prothrombin time, sodium citrate (0.105 M) was added immediately to the blood sample at a ratio of 1:10 (v/v), and values were recorded with a Beckman Coulter ACL 9000 Coagulation Analyzer according to the manufacturer's instructions. For 2/3 partial hepatectomy, virgin or pregnant mice (16–18 days post-coitum unless otherwise specified) were anesthetized, and the median and left lateral lobes of the liver were removed, as described(Ben Moshe et al. 2007).

Measurement of cell size

For FACS analysis, hepatocytes were isolated as described previously(Pikarsky et al. 2004) and forward scatter values were used as an indicator of cell size. To measure 'hepatocrit', equal numbers of isolated hepatocytes suspended in 50 µl of PBS were loaded onto a hematocrit capillary and centrifuged at 3000g for 10 min. The height of the hepatocyte column was divided by the total height. Mean hepatocyte volume was calculated by dividing the hepatocyte volume by the number of hepatocytes.

Antibodies

Primary antibodies against the following proteins and chemicals were used: BrdU (cat# MS-1058) from Thermo Scientific; Akt (cat# 9272), phosphoAkt Thr308 (cat# 9275), phosphoAkt Ser473 (cat# 4058), phosphoS6-kinase Thr389 (cat# 9234), phospho-4E-BP1 Ser65 (cat# 9451), and phospho-4E-BP1 Thr37/46 (cat# 2855), all from Cell Signaling; E-cadherin (cat# 610182) and p27 (cat # 610241) from Becton-Dickinsontransduction Laboratories; tubulin (cat# T9026) from Sigma; P53 (CM5) from Novocastra Laboratories and p21 (cat #6246) from Santa-Cruz biotechnologies.

Proliferation index

The percentage of BrdU-positive hepatocyte nuclei was assessed using the Kisight module of the Ariol SL 50® automated scanning microscope and image analysis system, according to the manufacturer's instructions. The same gating parameters were used for all sections. Ten fields in each liver were scored and the average percentage was calculated.

MRI analysis

MRI was performed on a horizontal 4.7T Biospec spectrometer (Bruker Medical), using a 3.5 cm birdcage coil. Mice were anesthetized (30 mg/kg pentobarbital, i.p.) and placed supine with the liver located at the center of the coil. Liver volumes were determined from multi-slice coronal and axial T1-weighted fast spin-echo images covering the entire liver (repetition time, 400 ms; echo time, 18 ms; slice thickness, 1 mm; field of view, 5 cm (coronal) and 3.4 cm (axial) using a 256 × 256 matrix). An observer who was blinded to the treatment group outlined the liver boundaries visualized in each slice, using image processing software (NIH image). To convert the pixel count to an area it was multiplied by the factor [(field of view)2 ×(matrix)2]. Total liver volume was calculated as the summed area of all slices, multiplied by the slice thickness. The post-hepatectomy liver volume of each mouse was expressed as a percentage of the preoperative volume(Ben Moshe et al. 2007).

Western blot analysis

Liver samples were homogenized in cell culture lysis reagent (Promega) with a Polytron homogenizer. Tissue lysates containing 100 g protein were separated by 12% SDS–PAGE, and assessed by Western blot analysis by means of sequential probing with the relevant primary antibody and a relevant anti-IgG conjugated to horseradish peroxidase (Jackson ImmunoResearch).

Immunohistochemistry and immunofluorescence

For BrdU immunostaining, sections (5M) were dewaxed and hydrated through graded ethanols, cooked in 25 mM citrate buffer pH 6.0 in a pressure cooker at 115 ºC for 3 min (decloaking chamber, Biocare Medical), and then transferred to boiling deionized water and allowed to cool for 20 min. After treatment for 5 min in 3% H2O2, slides were incubated with mouse monoclonal anti-BrdU antibodies diluted 1:200 in CAS-Block (Zymed) overnight at 4 ºC, washed three times with Optimax (BioGenex-HK583), incubated for 30 min with anti-mouse Envision+ K4007 (Dakocytomation), and developed with 3,3′-diaminobenzidine (Dakocytomation) for 15 min.

For measurement of E-cadherin immunofluorescence, sections (5M) were dewaxed and hydrated through graded ethanols, cooked in 10 mM Tris/0.5 mM EGTA at pH 9.0 in a pressure cooker at 115 ºC for 3 min (decloaking chamber), and then transferred to boiling deionized water and allowed to cool for 20 min. Slides were then incubated with mouse monoclonal anti-E cadherin antibodies diluted 1:50 in CAS-Block (Zymed) overnight at 4 oC, and revealed with Cy5-labeled secondary antibodies. For triple staining, the same antigen retrieval procedure was employed and the relevant primary and secondary antibodies were added.

Supplementary information references:

Ben Moshe, T., Barash, H., Kang, T.B., Kim, J.C., Kovalenko, A., Gross, E., Schuchmann, M., Abramovitch, R., Galun, E., and Wallach, D. 2007. Role of caspase-8 in hepatocyte response to infection and injury in mice. Hepatology (Baltimore, Md45(4): 1014-1024.

Pikarsky, E., Porat, R.M., Stein, I., Abramovitch, R., Amit, S., Kasem, S., Gutkovich-Pyest, E., Urieli-Shoval, S., Galun, E., and Ben-Neriah, Y. 2004. NF-kappaB functions as a tumour promoter in inflammation-associated cancer. Nature431(7007): 461-466.

Shiotani, M., Noda, Y., and Mori, T. 1993. Embryo-dependent induction of uterine receptivity assessed by an in vitro model of implantation in mice. Biol Reprod49(4): 794-801.